Thomas Kirsche

A client/server architecture for distributed workflow management systems

A specific task of distributed and parallel information systems is workflow management, In particular, workflow management system execute business processes that run on top of distributed and parallel information systems. Parallelism is due to performance requirements and involves data and applications that are spread across a heterogeneous, distributed computing environment. Heterogeneity and distribution of the underlying computing infrastructure should be made transparent in order to alleviate programming and use. We introduce an implementation architecture for workflow management systems that meets best these requirements. Scalability (through transparent parallelism) and transparency with respect to distribution and heterogeneity are the major characteristics of this architecture. A generic client/server class library in an object-oriented environment demonstrates the feasibility of the approach taken.<<ETX>>

Research: Communication support for cooperative work

Desktop conferencing and distributed graphical editing applications allow people to perform cooperative work regardless of their geographic location. One such example is the CoDraft application presented here. This collaborative system allows concurrent sketching on a shared drawing board. The current version is implemented on top of point-to-point communication services. It is shown that these services are not well suited for cooperative applications. Therefore, a multiparty communication platform is proposed that manages groups of application instances and multicast messages, handles group voting and transfers files to more than one target site at a time. This platform greatly simplifies the development of cooperative applications. Additionally, the platform allows the lower-level multicast functionality to be more fully exploited, thus improving response time and throughput performance.

Cooperative problem solving using database conversations

Cooperative problem solving is a joint style of producing and consuming data. Unfortunately, most database mechanisms developed so far; are more appropriate for competitive usage than for a cooperative working style. They mostly adopt an operational point of view which binds data to applications. Data-oriented mechanisms like check-in/out avoid this binding but do not improve synchronization towards concurrent usage of data. Conversations are an application-independent, tight framework for jointly modifying common data. The idea is to create transaction-spanning conversational working stages that organize different contributions instead of serializing accesses. To illustrate the conversation concept, an extended query language with conversational operations is presented.<<ETX>>

Building a real data warehouse for market research

This paper reflects the results of the evaluation phase of building a data production system for the retail research division of the GfK, Europes largest market research company. The application specific requirements like end-user needs or data volume are very different from data warehouses discussed in the literature, making it a real data warehouse. In a case study, these requirements are compared with state-of-the-art solutions offered by leading software vendors. Each of the common architectures (MOLAP, ROLAP, HOLAP) was represented by a product. The result of this comparison is that all systems have to be massively tailored to GfKs needs, especially to cope with meta data management or the maintenance of aggregations.

Architectural Issues of Distributed Workflow Management Systems

A specific task of distributed and parallel Information Systems is workflow management. In particular, workflow management systems execute business processes that run on top of distributed and parallel Information Systems. Parallelism is due to performance requirements and involves data and applications that are spread across a heterogeneous, distributed computing environment. Heterogeneity and distribution of the underlying computing infrastructure should be made transparent in order to alleviate programming and use. We introduce an implementation architecture for workflow management systems that meets best these requirements. Scalability (through transparent parallelism) and transparency with respect to distribution and heterogeneity are the major characteristics of this architecture. A generic client/server class library in an object-oriented environment demonstrates the feasibility of the approach.

Datenbankunterstützung für kooperative Gruppenarbeit

Das konventionelle Einsatzgebiet von Datenbanksystemen ist der administrative Bereich. Sein Anforderungsprofil ist bestimmt durch, eine zwar große Anzahl von Transaktionen, welche in der Regel aber nur von kurzer Dauer sind, da sie nur wenige Daten berühren. Wir stellen ein neues Konzept eines Datenbanksystems vor, das zur Unterstützung von kooperativer Gruppenarbeit mit langandauernden Verarbeitungszyklen geeignet ist. Der Ansatz basiert auf der Erweiterung eines herkömmlichen Datenbanksystems ( Zusatzebenenarchitektur). Diese Zusatzebene erlaubt die Definition von Datenqualitäten, Verhandlungsmechanismen und Benachrichtigungstriggern und dient so der Unterstützung kooperativer Gruppenarbeit. Empirische Ergebnisse aus einem Projekt mit einem süddeutschen A utomobilhersteller wurden bei der Konzeption des hier vorgestellten Systems berücksichtigt.

Functionality and architecture of a cooperative database system: a vision

A database system fostering the cooperative usage and modification of a common data pool should provide standard database functionality (e.g. application-independent correctness criteria and data modelling) plus means for a step-wise, cooperative refinement of data over a long period of time. Key ingredients are a hierarchical organization of work, a sound data model covering cooperative uncertainly, and support for long-living cooperative processes. Furthermore, mechanisms for data passing and hiding, negotiation means, and notification are of prominent importance. In the paper, the rationale behind such a functionality is described. Furthermore, a proposal is made for the software architecture of what is called a cooperative database system (CDBMS).

CoDraft: An Object-Based Drafting Tool on Top of Group Communication Services

Over the past few years, cooperation has become a key topic in computer science. This paper presents a cooperation tool named CoDraft and details the problems in developing it on top of today’s communication services. Then, a group communication platform tailored to the needs of cooperative applications is presented. Finally, an outlook on the implementation of CoDraft on top of our group communication platform is given.

Eine Datenanfragesprache für den praktischen Umgang mit vorläufigen Daten in und zwischen eng kooperierenden Gruppen

Konventionelle Datenbanksysteme unterstutzen kooperative Arbeit nur unzureichend. Defizite ergeben sich im Umgang mit vorlaufigen Daten (Datenreprasentation) und durch das klassische Transaktionskonzept (Datenverarbeitung). Die Gultigkeit der Daten wird durch committed und uncommitted festgelegt. Das Commit selbst tragt dabei die Bedeutung einer Freigabe und einer Festlegung gleichzeitig. Lange andauerende Entwurfszyklen, die die konsistenzerhaltende Auflosung der Isolation benotigen, sind nicht realisiert. Problemlosezyklen (problem solving activities, PSAs) legen einen Kooperationskontext fest, der lange dauern kann, konsistenzerhaltend ist und die Freigabe- und Festlegungsbedeutung des Commit entkoppelt. In diesem Beitrag wird das PSA-Daten- und -Verarbeitungsmodell vorgestellt, das die Kooperation in und zwischen eng kooperierenden Gruppen mit Hilfe der Datenanfragesprache CSQL (Cooperative SQL) ermoglicht. Ziel bei der Konzeption ist eine moglichst weitreichende Systemunterstutzung fur Datenbankbenutzer und -anwendungen.